Anachraphobia in upgrade tests

[Chris-Hibbert]

Describe the bug

in packages/inter-protocol, test -m test/swingsetTests/**/test-*.js provokes anachrophobia in #8157 at this commit.

For now, I'm going to disable these tests. They're old-style upgrade tests that we will need to recapitulate as docker-styole deployment tests before releasing, so there's nothing we need to preserve about them.

To Reproduce

Steps to reproduce the behavior:

1. Go to '...'
2. Click on '....'
3. Scroll down to '....'
4. See error

Expected behavior

User code should never provoke anachrophobia.

Platform Environment

Seems general. @warner reported:

I was able to reproduce locally (N=2). I changed the test to use defaultManagerType: 'xsnap',, and then I got a different error (again, twice): Error#1: resolution of kp124 is still pending.
if I had to guess, maybe this is our old friend "GC under V8 is not deterministic"
...
node-subprocess gets the anachrophobia too (N=2)
...
creationOptions = { nodeOptions: ['--jitless', '--gc-global', '--single-threaded-gc'] }, gets me anachrophobia too (edited) 

Additional context

Slack discussion

[FUDCo]

@Chris-Hibbert I suspect the command you gave there got mangled somehow on its way to being entered into the issue, because it's not syntactically valid, and the various obvious tweaks to it that I tried don't lead to anything useful or that runs.
If I just do yarn test in inter-protocol everything works fine for me, so I think there's more to the story.

[Chris-Hibbert]

I tried it again using yarn test test/swingsetTests/**/test-*.js, and got three failures, all reporting anachrophobia. Sorry about the mistake. You can also see it in CI here.

[FUDCo]

OK, I reproduced it. It's definitely connected to something in your PR, since it doesn't happen with master but does with the PR branch. That gives me a place to start digging...

[Chris-Hibbert]

We decided that the anachrophobia issue was a mere side-effect of a more relevant problem in #8157. No action required.